trixn86
Which is tragic because it is a neoclassical fiction book misleading millions of students. Anyone who is really interested in the history of economic thought will avoid that book for anything other than being a reference of the theoretical position of the neoclassical synthesis.
> I recognize that crafting a 'perfect' prompt is impossible, but neither is writing the perfect code, or even the 'perfect' code style guide
This is not about "perfect code" or style guides. For code styles we have linters, that does not require any instructions. I'm talking about software architecture for large complex web applications that dozens of people (that also change frequently) contribute to and that is spread across many different repos/services and an environment with constantly changing business requirements.
The terms that are relevant here are: coupling, cohesion, vertical slices, module boundaries and so on. This is software architecture and coding agents are terrible at this and there is no prompt and no instruction fixes this that could reasonably fit into a context window. I don't even think that coding agents fail because the lack the instructions and therefore adding instructions won't fix these failure modes.
There is no system prompt that can prevent this. If all you needed was some kind of rather short instruction we'd already have that baked into harnesses a long time ago. The problem is not that agents do not have the right instructions, they theoretically have everything there is to know about software architecture already in their training data.
nobody claimed otherwise
> Studies have shown that the vape smoke stimulates the vagus nerve which causes changes in the cardiovascular system that could likely cause heart disease
Can you name this study? Are you talking about "E-cigarettes and their lone constituents induce cardiac arrhythmia and conduction defects in mice"? This study did not find what you claim. All it found is that in mice the heart rhythm during and right after exposure increased. This is far from claiming that it "could likely cause heart disease" in humans. Also human switching studies point the other way. smokers who move to vaping show improvements in blood vessel function within weeks. I don't know why people really need to twist study results to depict vaping more harmful than it is. Yes, it is not healthy and you shouldn't do it. But all scientific evidence conclusively points to vaping, even heavy vaping, still being way less harmful than smoking because we already know that basically all harm comes from inhaling the residues of combustion, not from the nicotine. Condeming vaping as being equally bad as smoking is irresponsible and unscientific and prevents proven harm reduction for heavy smokers.
Also I don't think that there are many people that overdose nicotine regularily because it causes nausea and dizziness so people are normally self-regulating the nicotine dose automatically. So the scenario you are proposing doesn't seem to be something we can actually observe as a common consumption pattern.
Too much nicotine is far closer to drinking too much coffee than to smoking and has similar acute effects which is why almost nobody does it on a regular basis.
It was harder to design Symmetric Encryption using e.g. AES-256 than to break it? As far as I know it's design is pretty straight-forward but there is no known way to break it even with all compute of the planet at your disposal (excluding trivial ways like exposing the key).
I never said this. All I said is we don't have the conversation and therefore we can't determine how easy or hard of a problem it was.
> And it's very possible Terra or GLM could crack it, turn off their web access and try yourself.
I'm questioning why this should be labeled "Astra" breaking anything implying it required "the best" model to do it when in fact any other half-decent model might have been able to do this as well.
EDIT: Okay seems like the actual prompt is published, just not on the same article that was linked. Maybe I'll give it a try.
Also: Why should we assume Terra, GLM or any other less SOTA and less expensive model wouldn't have been able to do the same?
Semantic class names make no sense in a component context and that's why tailwind feels like a liberation in the context of component-based web applications. A component and its interface is the semantic meaning, the markup and styles it contains are scoped to that component and should not be / need not be semantic at all. This greatly reduces cognitive load.
The main issue with mainstream economics is that it is some cargo-cult fairytale of a world that doesn't exist where everybody individually behaves rationally and that is supposed to lead to a desired outcome overall. But this is simply a fallacy of composition because individual behavior (especially if it involves spending/saving/investment decisions) is always liked to others through balance sheets, basically simple accounting.
A simple illustration of the fallacy of composition: One person can stand up in the cinema to improve their view. But it would be a mistake to think that therefore if everybody in the cinema stood up everybody improved their view. In fact already the first person standing up, while improving their view, did impair the view of the person behind them. This is the fallacy of composition and when it comes to saving money it's known as the "paradox of thrift" which states that an economy as a whole can not save any money.
Saving by definition means that you spend less than your income in any given period which necessarily requires one or more other entities to spend more than their income to make the math work. Therefore savings and debt are the same thing, the two sides of the same coin.
Instead of just looking at a single individual you can divide the economy into sectors and watch the flows between those sectors which always have to add up to 0 (this is called sectoral balances or stock-flow-consistent modeling). If we take a simple closed economy (so not considering the government and there is no foreign sector) and divide it into households and companies and we also assume that the household sector as a whole wants to net save 5% of their income in every period (which is empirically about correct) that means to keep income for the households steady the companies have to run a deficit of equal size in every period (i.e. spend more than they earn). What that means is that savings always and necessarily equal investment which is usually stated as I = S. But if households try to save more than the companies expected and which will lead to less sales than expected companies are forced into a deficit in p1 which they will likely react to by decreasing their investment in the following period p2 even when the interest rate dropped because any investment is pointless if there is no demand for the products to invest in producing. So the propensity of companies to invest is mainly influenced by their expectations of future sales and this is genuinely uncertain. If companies do not invest at least as much as households save the economy will inevitably shrink and both income and investment will drop together in the following period. Of course when we consider a government it could also run a deficit to compensate for the gap (the US would be a prime example for that) and/or a country could run a trade surplus (Germany, China).
The main issue is that other than neoclassical mainstream economists think we are not living in a world where there is only Robinson Crusoe on a lonely island that knows when he saves some fish it's because he wants to manufacture a fishing rod. But that is essentially what neoclassical models model, they pretend that everybody is basically acting like a hive mind that knows all future spending and saving decisions in the future and where money is merely an infrastructure to facilitate barter. This is entirely wrong. Money is non-neutral and it is not just another commodity. It is created when somebody incurs a debt (e.g. by taking a loan from a bank) and it is destroyed when the debt is repaid. Banks create money, they are not acting as intermediaries between savers and borrowers. This has long been argued by Post-Keynesians and is well known for at least 100 years but mainstream economists only admitted that publicly just a few years ago.
See "Money creation in the modern economy" by the Bank of England (2014) for an example of that recognition.
Closing with to quotes:
"I have found out what economics is; it is the science of confusing stocks with flows" - Michal Kalecki
“it is much more realistic to say that banks "create credit", that is that they create deposits in their act of lending, than to say that they lend the deposits entrusted to them. And the reason for insisting on this is that depositors should not be invested with the insignia of a role they do not play. The theory to which economists have clung so tenaciously […] attributes to them an influence on the 'supply of credit' which they do not have. " - Joseph Schumpeter
To illustrate that it's often helpful to think in extreme scenarios. Imagine every household starts to save 100% of its monetary income. What does that mean? It means that nothing is sold anymore and companies have 0 revenue which will soon lead to a complete collapse of the economy and everybody becoming unemployed and loosing all their income as well unless the companies will take on debt to keep paying the wages/profits (which they will not do when there is no demand for their products).
Money needs to be spent or else demand will drop and the economy can enter a vicious downward spiral (a deflationary collapse / debt deflation). The most impressive example of that was the great depression.
If some sector of the economy wants to net save (usually those are the households) to keep the same level of economic activity (and therefore jobs and income) somebody else needs to spend money they don't have, i.e. they need to go into debt.
The main issue is that in a society with division of labor there is no mechanism that keeps saving and investing in line so that employment and income is kept on a stable level. The mainstream neoclassical economic theory claims that the interest rate is always and automatically making sure that for every dollar saved someone else will invest it but this is based on the assumption that investors have infinite and complete knowledge about what everybody else will do in the future and that the economy will always and necessarily tend towards an equilibrium state of full employment. They are obsessed with "equilibrium" which is why in mainstream publications you will find that word everywhere. But in reality the economy is a non-equilibrium complex system with pro-cyclical feedback loops and all the interesting characteristics worth studying are non-equilibrium behaviors of the system.
Some recommended literature regarding that topic:
- The two essays "What is money?" and "The credit theory of money" by Alfred Mitchell-Innes - "The theory of economic development: an inquiry into profits, capital, credit, interest, and the business cycle" by Joseph Schumpeter - "Debunking economics" by Steve Keen - "Can it happen again?" by Hyman Minsky - "Debt: The first 5000 years" by David Graeber
Money is credit. It's not an asset. Gold or Bitcoin are not money, they are an asset. The economy is credit-based, it's not a barter economy.
Is it just me or is everybody else just as fed up with always the same AI tropes?
I've reached a point where I just close the tab the moment I read a headline "The problem". At least use tropes.fyi please
The only context I consistently found to be useful is about project-specific tool calling. Trying to provide natural language context about the project itself always proved to be ambiguous, inaccurate and out-of-date. Agents are very good at reading code and code is the best way to express context unambiguously.